Re: ambigious 'top' results

2008-12-16 Thread Mel 
On Tuesday 16 December 2008 16:25:45 KES wrote:

>
> ===>>>   4405 prcfr  1999 cpu0:
> time 34 dtbuf 6486 totfr

Time counter. It it supposed to generate kern.clockrate interrupts per CPU.
-- 
Mel

Problem with today's modular software: they start with the modules
and never get to the software part.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: ambigious 'top' results

2008-12-16 Thread Mel 
On Tuesday 16 December 2008 16:31:35 Wojciech Puchar wrote:
> are you using 7-branch?
>
> in my 7.1 system there is same problem. top shows often nonsense in CPU
> usage of a process, but totals are OK.

there's 4.3% system, so use top -S to show system processes.

> you even gave an example - bzip2 can't use 0% CPU running - more close to
> 100% :)

If bzip2 is blocked, it'll be running at 0%. Example:
bzip2 -cf - /path/to/large/file | (sleep 900; cat - > /dev/null)

-- 
Mel

Problem with today's modular software: they start with the modules
and never get to the software part.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: ambigious 'top' results

2008-12-16 Thread Paul B. Mahol 
On 12/16/08, KES  wrote:
> Здравствуйте, Questions.
>
> My system is overloaded, I run top to figure out which precess take
> time. In both cases CPU/WCPU I do not see which process take all the
> time. Which process take all processor time??
>
> last pid: 24535;  load averages:  3.86,  3.07,  3.07up 1+13:34:28

If last pid grow very fast, than some process are spawing like mad.

-- 
Paul
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: ambigious 'top' results

2008-12-16 Thread Wojciech Puchar 

are you using 7-branch?

in my 7.1 system there is same problem. top shows often nonsense in CPU 
usage of a process, but totals are OK.


you even gave an example - bzip2 can't use 0% CPU running - more close to 
100% :)



as long as everything goes fine 100% CPU load is OK. if it's mostly mysql 
then optimize your database, if you can't change hardware

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

ambigious 'top' results

2008-12-16 Thread KES 
Здравствуйте, Questions.

My system is overloaded, I run top to figure out which precess take
time. In both cases CPU/WCPU I do not see which process take all the
time. Which process take all processor time??

last pid: 24535;  load averages:  3.86,  3.07,  3.07up 1+13:34:28  17:25:19
193 processes: 5 running, 187 sleeping, 1 stopped
CPU: 95.1% user,  0.0% nice,  4.9% system,  0.0% interrupt,  0.0% idle
Mem: 172M Active, 171M Inact, 97M Wired, 23M Cache, 60M Buf, 25M Free
Swap: 2048M Total, 115M Used, 1933M Free, 5% Inuse

  PID USERNAMETHR PRI NICE   SIZERES STATETIME   WCPU COMMAND
 2942 mysql42   40 52412K 22132K sbwait   0:00 22.80% mysqld
88534 root  1 -580  3288K   616K bpf  1:22  0.00% ipguard
89345 root  1 -580  3288K   632K bpf  1:20  0.00% ipguard
  213 root  1  440  3180K   472K select   0:55  0.00% syslogd
  709 bind  4  440 58480K 23996K select   0:42  0.00% named
 4719 pgsql 1  440 42508K  1620K select   0:18  0.00% postgres


last pid:  2982;  load averages:  2.59,  3.37,  3.35up 1+13:23:39  17:14:30
196 processes: 5 running, 190 sleeping, 1 stopped
CPU: 94.8% user,  0.0% nice,  3.4% system,  0.0% interrupt,  1.9% idle
Mem: 129M Active, 168M Inact, 112M Wired, 11M Cache, 60M Buf, 68M Free
Swap: 2048M Total, 119M Used, 1929M Free, 5% Inuse, 8K In

  PID USERNAMETHR PRI NICE   SIZERES STATETIMECPU COMMAND
 2942 mysql41   40 50748K 21240K sbwait   0:00 13.23% mysqld
 2451 root  1  440  3532K  1924K RUN  0:01  0.10% top
 2254 root  1  960 11380K  6988K RUN  3:12  0.00% bzip2
66162 root  1  440  9192K  2396K RUN  1:26  0.00% verlihub
88534 root  1 -580  3288K   616K bpf  1:21  0.00% ipguard
89345 root  1 -580  3288K   632K bpf  1:19  0.00% ipguard
  213 root  1  440  3180K   472K select   0:54  0.00% syslogd
  709 bind  4  440 58480K 23036K select   0:42  0.00% named
 4719 pgsql 1  440 42508K  1620K select   0:18  0.00% postgres


#systat -v
4 usersLoad  2.71  2.75  3.02  16 дек 17:21

Mem:KBREALVIRTUAL   VN PAGER   SWAP PAGER
Tot   Share  TotShareFree   in   out in   out
Act  136916   11584   79968431004   29768  count
All  190728   13620  297186863628  pages
Proc:Interrupts
  r   p   d   s   w   Csw  Trp  Sys  Int  Sof  Flt   1656 cow2003 total
  1   5 186   11k 7340  75k4  35k 7189   4198 zfodatkbd0 1
  232 ozfod   sio1 irq3
54.6%Sys   0.0%Intr 45.4%User  0.0%Nice  0.0%Idle   5%ozfod   sio0 irq4
|||||||||||   daefr 4 ata0 irq14
===>>>   4405 prcfr  1999 cpu0: time
34 dtbuf 6486 totfr
Namei Name-cache   Dir-cache 35373 desvn  react
   Callshits   %hits   % 18913 numvn  pdwak
96679420  97  47   0  8279 frevn  pdpgs
  intrn
Disks   ad0   ad1  102212 wire
KB/t   0.00 15.67  136888 act
tps   0 5  230508 inact
MB/s   0.00  0.07   24388 cache
%busy 0 05384 free



  --
С уважением,
 KES  mailto:kes-...@yandex.ru

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"